Weekly
Get a weekly written recap of your Mixpanel metrics
Each Monday you get a few paragraphs on what went up, what went down and what stands out.
- Runs
- Weekly
- Apps
- Mixpanel
- AI
- Yes, uses the built-in AI allowance
- Access
- service account
Send this to your assistant
Create a ClickRuntime automation: every Monday at 9am, pull last week's sign-ups, active users and checkout events from Mixpanel and email me a short summary of what changed.
First time? Adding ClickRuntime to your assistant takes about two minutes.
Why use it
Why teams use this automation
Most product managers can read a Mixpanel report but do not have time to open six of them each Monday and write down what moved. The numbers sit in separate charts for retention, funnel conversion and event volume, and the story of the week, such as onboarding improved after the new checklist while invites fell, is never written anywhere. Leadership asks what changed, and the answer is rebuilt from memory.
Once a week ClickRuntime runs a set of Mixpanel queries for the main metrics, hands the results to its built-in AI to write a short plain-English narrative of what rose, what fell and what looks unusual, and emails it. No OpenAI or Anthropic key is needed, because the AI step is part of ClickRuntime.
It works well as the first paragraph of a weekly product update.
What you get out of it
The story of the week is written down
Instead of raw tables, you get sentences like conversion from trial to paid slipped while invitations grew. A written record means the next quarter's review can look back at an actual narrative, not scattered charts.
Leaders get answers without an analyst
An executive can read three short paragraphs on Monday morning and ask informed questions. The product manager spends the saved hour on the follow-up, not on compiling screenshots into a slide.
Quiet metrics get noticed too
Charts nobody opens, like a slowly declining invite rate, are queried every week and described. The summary can mention changes that would never have been spotted by someone only watching the headline number.
One consistent set of queries
The same queries run every week, so comparisons between weeks are fair. You avoid the common problem where this week's slide used a different date range or segment than last week's.
Faster weekly planning
Because the narrative arrives before planning, the team starts Monday with a shared view of what shifted. Priorities for the sprint can respond to real movement rather than to whoever remembers the loudest chart.
How it works
What happens, step by step
Every Monday, ClickRuntime pulls last week's key numbers from Mixpanel, such as sign-ups, active users and checkouts.
AI reads them next to the week before and writes a few short paragraphs on what went up, what went down and what stands out.
The write-up arrives in your inbox as a quick read to start the week.
Before it runs live
- Connect Mixpanel
What you get
Build your own
Need it to work differently?
Your own rules
Conditions, thresholds, filters and exceptions, written the way you'd explain them to a colleague.
Any number of apps
One automation can read from one app and update several others. Over 248 apps are ready, and anything with an API works too.
AI where it helps
Summaries, classification, extraction and drafts use ClickRuntime's built-in AI, with no extra key.
On a schedule or on each event
Runs as often as every 5 minutes, or the moment an app sends an event, and keeps track of what it already handled.
Changes you can ask for
- Run it at a different time, more or less often, or only on weekdays
- Change what the AI looks for, or how it writes the result
- Add another app, for example also send it to Slack
- Combine it with other Mixpanel steps into one automation
Under the hood
The code that runs this automation
from datetime import date, timedelta
from clickruntime import on_trigger, credentials, http, ai, notify
PROJECT_ID = 2195193
EVENTS = ["Sign Up", "$session_start", "Checkout Completed"]
@on_trigger
def run(event):
auth = ("ops-reader.mp-service-account", credentials.secret("MIXPANEL_SERVICE_SECRET"))
end = date.today() - timedelta(days=1)
lines = []
for ev in EVENTS:
data = http.get("https://mixpanel.com/api/query/segmentation", auth=auth, params={
"project_id": PROJECT_ID, "event": ev, "unit": "week",
"from_date": str(end - timedelta(days=13)), "to_date": str(end)}).json()
weeks = sorted(data["data"]["values"].get(ev, {}).items())
lines.append(f"{ev}: " + ", ".join(f"week of {k}: {v}" for k, v in weeks))
summary = ai.generate("\n".join(lines), max_output_tokens=400,
system="Summarise in 4-6 short sentences what changed week over week and what stands out.").text
notify.email_me("Mixpanel: last week in short", summary + "\n\nRaw numbers:\n" + "\n".join(lines))
return {"events": len(lines)}FAQ
Common questions
Does the AI invent numbers?
Which metrics does the summary cover?
Can I send it to a channel rather than email?
What permissions does it need?
Does it need AI?
How many runs does it use?
Can I change how it works?
Related
You might also like
Get yesterday's Mixpanel KPIs by email
Every morning you see daily active users and sign-ups next to the same day last week.
See how it worksGet a Slack alert when sign-ups fall behind
If sign-ups are running 30% under their usual pace, you find out within the hour.
See how it worksGet a daily news digest in Slack
Each morning your team gets a short summary of what's new on the sites and blogs you follow.
See how it worksKeep exploring
Set up “Get a weekly written recap of your Mixpanel metrics” from your assistant
Copy the prompt above, or connect your assistant first. Free plan, no credit card.